This subproject is one of many research subprojects utilizing the resources provided by a Center grant funded by NIH/NCRR. Primary support for the subproject and the subproject's principal investigator may have been provided by other sources, including other NIH sources. The Total Cost listed for the subproject likely represents the estimated amount of Center infrastructure utilized by the subproject, not direct funding provided by the NCRR grant to the subproject or subproject staff. Almost 15 years have past since diffusion tensor imaging was introduced. During this time period, data acquisition technology, both in terms of hardware and pulse sequences, has developed with a considerable speed. Currently, DTI is built in many scanners as a part of commercial packages and DT images with a decent image resolution (typically 2.5 mm isotropic resolution) and quality can be acquired within 10 min. However, there remain many "bottlenecks" that hamper the application of this promising technology to actual clinical and research studies. In this proposal, we identify four relevant bottlenecks in data acquisition and processing fields and concentrate our resources on these topics. These are;1) improvement of image resolution / scanning time relationship (Aim 1), 2) development of data acquisition and processing tools for tract-specific multi-modal examination (Aim 2), 3) development of voxel-based quantification tools (Aim 3), and 4) development of software that makes state-of-the- art DTI technology available to wider ranges of researchers (Aim 4). Aim 1: High-resolution and rapid DTI for the brain and spinal cord imaging This aim is closely related to our overall goals mentioned in the Overall Resource Description and Goals: study of brain and spinal cord of pediatric populations. A short scanning time is essential for young populations who may not be cooperative enough. The study of the spinal cord is a further challenge because of its small size and increased motion. Based on the fact that resolution of DTI is limited by SNR and echo-train length of EPI data acquisition, we will push the boundary by employing higher field (7T) and multi-channel parallel data acquisition. Aim 2: Development of tools for tract-specific multi-modal examination of the white matter and application to brain development We will use DTI as a tool to provide an anatomical template of the white matter and, by combining with various quantitative MRI approaches (relaxation rate measurements, magnetization transfer measurement), we will establish a fast protocol to examine status of individual white matter tracts. This includes acquisition of co- registered multi-modal data and development of techniques to evaluate status of individual white matter tracts. Together with our coinvestigators in TRD2 and TRD4 and our collaborators, we will first use these tools to characterize normal brain development using our normal pediatric database and then apply them to the specific diseases. Aim 3: Development of voxel-based morphometry and application to brain development This aim is important because quantification of tensors is different from that of scalar images and quantification tools for DTI are not yet well established. This will be a collaboration combined effort with TRD 4. We will perform 3 steps. First, we will investigate voxel-based morphometry tools based on Affine and LDDMM transformation. Second, we will develop landmark-based DTI registration method. Third, we will develop and evaluate direct tensor- to-tensor registration technology. Results of conventional scalar-based registration (first step) and the tensor- based registration (third step) will be compared and evaluated for improvement of registration quality. The second landmark-based method is important as a gold standard to evaluate automated approaches and also for registration of severely pathological brains for which automated methods fail. As a first target of the application studies, we will apply these tools to characterize normal brain development followed by assessment of the different pathologies studied by our collaborators. Aim 4: Development and support of user-friendly DTI data processing and analysis tools Our DTI data processing software developed as part of the Resource, DtiStudio, now has more than 300 registered sites. We will keep improving the software. In this grant period, the following functions will be added or improved; 1) improvement of existing functions such as I/O functions, ROI drawing function, visualization, 2) development of landmark-based B0 distortion correction tool, and 2) incorporation of new functions developed under Aim 2 and Aim 3.